{"id":"W4402671446","doi":"10.18653/v1/2024.arabicnlp-1.27","title":"Dallah: A Dialect-Aware Multimodal Large Language Model for Arabic","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Arabic; Computer science; Natural language processing; Linguistics; Artificial intelligence; Speech recognition; Philosophy","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003854118,0.0007300754,0.000343684,0.0003203462,0.0004249138,0.0009297042,0.001042715,0.0005428378,0.007353027],"category_scores_gemma":[0.001401307,0.0002355024,0.0007010445,0.0002081703,0.0002529525,0.001049297,0.001091877,0.001060127,0.002945547],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006369521,"about_ca_system_score_gemma":0.0008277261,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00926529,"about_ca_topic_score_gemma":0.01718371,"domain_scores_codex":[0.9998072,0.00006758707,0.0000122329,0.00005755108,0.00003842106,0.00001706704],"domain_scores_gemma":[0.999749,0.0001095982,0.00001506624,0.00003400838,0.00006566373,0.00002665207],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001141805,0.000376569,0.003123429,0.0004915012,0.000278671,0.0007203669,0.001065545,0.3012132,0.04156466,0.03316758,0.06687428,0.5499824],"study_design_scores_gemma":[0.00003818091,0.00006564674,0.0002980843,0.00001604991,0.00002604959,0.00009605072,0.0000889158,0.9682236,0.004959862,0.006099181,0.02005737,0.00003090186],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.043565,0.0007346403,0.9103262,0.0007563105,0.0003032048,0.000317655,0.003150846,0.03156676,0.009279386],"genre_scores_gemma":[0.3820004,0.0005406424,0.5816019,0.0006258894,0.00009028168,0.0007509326,0.008182444,0.001902155,0.02430523],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00926529,"threshold_uncertainty_score":0.02459836,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01374668446138967,"score_gpt":0.304899655221933,"score_spread":0.2911529707605433,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}